Based on your feedback, we pursued a web experience for IE users consistent with what is available on iOS and Android devices – even where this meant we would be adding non-standard web platform features. We believe that this is a more pragmatic approach to running today’s less-standardised mobile web.
Thank you, web developers, for turning mobile Safari into the new Internet Explorer. Have you people learned nothing?
I remember spoofing OSX safari as IE to make problematic sites work better.
I have to say that as an old time Mac user this turn of events is both delicious and very pleasing
I guess this sort of stuff happens if you skate to where the puck is going to be – something Microsoft stopped doing over twenty years ago.
‘This turn of events’, meaning a non interoperable mobile web?
That pleases you?
That’s one way to look at it. Another is that, as always, the most powerful platform sets the standards, and standards are necessary. Of course some standards are good and some are not good.
The Microsoft web standards created by the dominance of IE were not good because the only reason Microsoft fought so hard to make their IE the dominant browser was to shut down the threat that they thought the browser posed to their domination of the PC world. Once Microsoft and IE had ‘won’ they immediately lost all interest in browser development, and IE entered a period of stagnation which unfortunately froze much of the web at the IE6 level, a very unfortunate standard indeed.
Whether Apple’s dominance of mobile browsing will lead to unfortunate effects (or indeed even last – look what happened to IE once it had ‘won’) is too early to say. However Apple has done much to move the mobile web forward so perhaps the period of Safari domination of mobile browsing will not cause too many problems.
Of course as an old Mac head the whole thing makes me grin broadly. Poor old Microsoft, they had it all and they let it all slide between their hands because after twenty years of basically defending their hegemony based on a short period of innovation back in the 1980s and early 1990s innovative product development was so far from their corporate DNA that they wouldn’t recognise innovation if it came up and bit their butt. Which it did, and then sank it’s teeth into their jugular.
Serious question: how old are you?
Old enough to remember how Microsoft established an insecure, ugly, dysfunctional and disempowering operating system as the world personal computing standard. They did it by being very clever during the most fluid phase at the dawn of the PC period, and then ruthless in defence of their position, and by recognising that the mass purchase of PCs would start in the workplace and by dominating corporate computing the consumer space would follow.
I watched a generation of users become used to wrestling with an unstable, insecure, needlessly complex and arcane operating system, and as a result a generation grew used to the notion that using a computer was complex, difficult and threatening. That using a PC required special knowledge and skills. That most users were bumbling amateurs who needed support from experts (hence IT depts loved Microsoft because their software empowered them and guaranteed their jobs and power).
So am old enough to truly appreciate and celebrate the fact that Microsoft have been cast into a tiny minority role in the new age of mobile device. That’s where they belong. It’s what they deserve.
MS totally dominated the business market before Windows was widely available. (DOS was still widely used in the early 90s).
Early version of Windows were not much more than a DOS shell. Windows NT,which underlies all modern versions of Windows, is a totally new OS.
I never mentioned Windows and none of what you said makes any difference to my comment unless you think using DOS was less complex, difficult and threatening for end users than using Windows?
I suggest you study some neuroscience. You may discover the interesting fact that pretty much every complex system is equally difficult. There is really no ‘hard’ or ‘easy’ way to perform any complicated task – merely different ways.
A system that has a steep initial learning curve may actually be much simpler when complex tasks are involved. When ‘glass cockpits'(screen based instruments) were first introduced to aviation they were considered easier and safer to use than traditional analogue instruments. However recent data suggests that ‘glass cockpits’ may be more difficult to use under certain critical (eg takeoff, landing go arounds) situations and cause a higher accident rate.
The Mongols have a saying – “horses cannot think like a person so we teach our children to think like a horse.” Likewise we should teach people to understand how a computer works rather than trying to make computers behave like a human.
Jolly interesting but irrelevant to the real world where Ms-Dos and Windows, up to and including Windows XP, were all unnecessarily complex and disempowering to end users. The reason it’s possible to say that is because an equally powerful operating system, the Mac OS, was available at that time which was much, much easier to use and far better designed. Eventually, with Windows 7, Windows got to be almost as easy to use as OSX (although still fairly ugly and dysfunctional in places) but then Microsoft in an act of vandalism, precipitated by the realisation that Windows was not suitable to leverage any sort of advantage in the new and giant mobile computing devices markets, released Windows 8 which was a huge step back in usability.
Why on earth do people have to know how a computer works in order to use it? That’s a ludicrous idea. I can drive a car, surf the web, operate a very sophisticated audio visual system, without knowing how it works. I just need to know how to operate those things and operating something very complex can, with very good design, be made simple.
There is no excuse for bad design which disempowers people. There is no excuse for ugliness or unnecessary complexity.
Check out which OS is the most commonly referenced here
http://www.interfacehallofshame.eu/www.iarchitect.com/shame.htm
Mac OS Classic, “far better designed”?…
And that site you linked to reflects the popularity of Windows.
Edited 2014-08-07 00:11 UTC
Okay.
Here^aEURTMs the thing. I got my first Macbook when I was 14. I remember the fun of joining the ^aEURoeApple community^aEUR and watching the original iPhone announcement with everyone on the RoughlyDrafted Forums and AppleInsider and bashing Microsoft and reading all the history that you apparently were old enough to witness.
Given a choice between Windows and OS X, I would pick OS X every time.
What I want to know though is, when do you grow out of that? When do you stop corporate cheerleading and swear allegiance to independent thought and just let the past go? I figured 16 was good enough time to exit *that* community, in part because I got utterly sick of 30 and 40 and 50 year old consumerist attitudes from otherwise sentient, rational and thinking beings whom the moment Apple or Microsoft (or Google or Adobe) did anything ^aEURoebig^aEUR would drop all of that and turn into unpaid representatives of whichever company they most recently threw the most amount of money at.
I taught myself to program and web design at a time when web standards were a respectable thing to adhere to not because it made websites more compatible with Safari or Firefox, but because when adhered to, they are the basis of browser independence and it stops being relevant whether the browser came from Apple or Microsoft or anybody else, so long as it respected those same standards.
I moved onto systems programming, but to see that the world has a new IE just makes me shudder. What happened to the web developers that actually bothered to pay attention to web standards rather than targeting specific browsers? What happened to the culture of browser independence and platform independence? What happened to pushing the envelope not just for one browser, but all browsers? Why is it that we^aEURTMre back to browsers pretending to be other browsers out of necessity again?
And why is it, that somebody that is probably 20 or more years older than I am and cares enough to comment on a post pertaining to web standards, accepting this? Not just accepting it, but smiling with schadenfreude at the idea of a particular company he or she doesn^aEURTMt like having to use a stupid hack for website compatibility because the web developer community apparently stopped caring about things like browser independence and cross platform support. Because of actions taken more than 20 years ago by a publicly traded for-profit company.
Isn^aEURTMt browser independence and the open web just as noble a goal when it is browser independence from WebKit instead of IE?
Slll,
That’s what I was thinking too. What IE did in the past was terrible for web standards, the bane of web developers everywhere. It took us a good decade to get away from that proprietary legacy mess. Now that a new browser is in a similar position with mobile, why should it be worthy of celebration?
It kind of reminds me of war – some people are so ingrained with “our side” versus “their side” mentality that their objectivity gets compromised. Right and wrong get reinterpreted with heavy bias based on the side committing the actions, rather than seeing things for what they are and giving everyone a fair assessment.
Edited 2014-08-02 13:46 UTC
WebKit is open source.
The IE render engine was closed.
Times change. For the better.
Tony Swash,
That’s completely unrelated to our discussion of standards. Having an open source WebKit is good, but that does not mean you should overlook the negatives of browser hacks to render websites correctly on all mobile devices. Now that IE will implement them just to make pages render correctly, it will significantly solidify the presence of non-compliant markup on the mobile web.
Except what made IE so bad wasn’t that it was closed, it’s that it broke standards and so many websites were reliant on those broken standards and browser identification to render correctly. This was a horrible practice, which forced Opera to lie about it’s identity to render pages correctly. While some things change (IE having a reversal of roles), it seems other things do not (the need to support proprietary features to render correctly).
Not everyone uses Webkit.
Lock-in to an open source rendering engine isn^aEURTMt any better than lock-in to a closed source one.
Different WebKit-based browsers use different versions and Chrome even forked WebKit into Blink. And using WebCore doesn^aEURTMt necessarily mean you^aEURTMre using JSC and using JSC doesn^aEURTMt necessarily mean you are using WebCore.
It^aEURTMs more complex than ^aEURoeWebKit is open source therefore it doesn^aEURTMt matter if we^aEURTMre locked in^aEUR. What matters is the Webdev community was moving towards browser independence and now Webdevs are targeting MobileSafari. If IE were open source and available on every single platform, IE lock-in would still have been a problem: it undercut user choice.
More important than that though, the lingering ^aEURoeus vs them^aEUR mentality undercuts objective thought. It^aEURTMs fun when you^aEURTMre a teenager and you^aEURTMre still exploring different systems of thought, but it just creates problems and is poisonous in any serious community.
Yes, and over the years, Blink and WebKit will accumulate more and more differences. The current situation when web engines by Apple and Google are very similar, is only temporary.
What happenned?
Mobile sites simply evolved from novelty and copetitive advantage of core web companies to a survival asset in any serious business.
This in turn onboarded hordes of opportunistic developers that took the path of lower resistance, getting rewarded greatly for delivering quick and dirty solutions to desperate corporate laggards.
Keeping track of ever changing, muddy waters of standard setting process, that is additionally a arena of covert corporate infighting takes a lot of energy and understanding beyond basic web dev techniques.
Oh please, the bullshit is thick with this one!
EVERY SINGLE THING that has EVER been attributed to MSFT’s “success” can be preceded by “And then the other guy did something REALLY fricking stupid” going all the way back to the owner of CP/M going flying instead of talking to IBM.
The rise of Windows? Apple fired the guy with taste for a series of stuffed suits, IBM burned the bridges with all of the OEMs with the MCA debacle so that when they came out with OS/2, a much faster and more stable OS, copies of OS/2 were seen as plague blankets by the OEMs and rejected. BeOS? First went for an oddball CPU by AT&T (Hobbit) and when that was canceled went for the overpriced and underpowered PPC, which meant they had to compete with a much larger corp for each chip, and by the time they realized “duh we should be on X86!” they had already bled through the cash.
The same can be said for everything else. IE? Some genius at netscape said “Hey we are ahead, ya know what we should do? Massive rewrite lulz”and came out with the abomination known as NS4 which had people rushing to IE in fricking droves because “hey at least IE can render 3 pages in a row without BSODing the entire system!”. MS Office? The braintrust at Wordperfect said “Hey we are making money hand over fist with WP, there is no need to make a usable port for this “Windows” thing, not when we can use this “GW thing” to just wrap our DOS program…its stable, right?”…/facepalm/
So if it makes you happy to buy into the delusion that Gates was Darth Vader? Go for it, I have a bridge you might be interested in BTW. But if you look at the actual facts one thing becomes perfectly clear…MSFT was handed the markets they were dominant in by competitors that were full retard.
THIS, this right here, is why after billions poured down the crapper MSFT has lower than the margin for error when it comes to phones and tablets, they simply have no experience competing with companies that aren’t suicidal stupid. Whether you like it or not Windows IS a good product, but they had 15 years with ZERO real competition so they had time to figure it out and even then we ended up with messes like Vista and 8, where they have real competition and hard deadlines? Surface RT and WinPhone 8 not being BC with WinPhone 7…yeah that was smart, “lets burn the early adopters when we have low single digits!”…/double facepalm/
Give the man a beer!
(and when BeOS was on x86, it was simply too late, after Win98 already came out, which was decent enough)
PS. I hope you have a template for those posts
Anything that excludes Microsoft pleases me
You should discuss that with your psychiatrist.
You shortsighted fool, this is a much bigger problem for FirefoxOS or Jollas of the world than to Microsoft.
What it means for the web is not good, but nobody ever claimed web devs where the sharpest knives in the drawer, the amount fo CPU power wasted these days to run all of that crappy javascript is just dumb, I’d much rather have an old school pure HTML site, even if it uses slightly more bandwidth to load, push back
against the ISPs, don’t cave.
However…
The sweet, sweet irony greatly pleases the former(Went all Linux in 2005) Mac user in me. That and who in their right mind would cry if IE finally got put out to pasture over this? At least Safari is built on Webkit, which is OSS.
Lets see…large corp with previous history of lock in uses extensions that don’t follow web standards and due to their sheer size causes everyone else to have to use workarounds and fixes to get around their near monopoly…yep all we have done is swap Redmond for Cupertino, wonderful.
Just going to quote myself here:
http://www.osnews.com/comments/26630
Sad that almost two years later the situation has only gotten worse.
The ones that turned webkit into IE6 are dumb web developers. Seems the new ones are just as stupid as the old ones that created the the “optimized for IE6” sites.
Rendering bugs and behaviour that is not in the spec should be reported as bugs not “optimized for”.
You never lose a moment to spread the word of the Redmond Bullshit Bible, do you?
Are you fucking retarded? Serious question, because that was a useless comment.
Let’s unpack for a second what you’re saying:
1. You blame web developers for taking advantage of the platform given to them, and then implicitly expect them to understand the nuances of the various standardizations efforts underway. Not going to happen.
2. You blame Microsoft for being pragmatic in the face of a broken web. A web that was broken by competing vendors, and who have not shown ANY willingness to solve the problem.
Apple OPPOSES the Touch Event W3C Recommendation. Opposes. There is no concerted effort to remedy this.
Microsoft had to implement this non standard behavior, because Apple (and the iPhone’s clout) have completely fucked the mobile web.
So no, go fuck yourself. I deal with this shit everyday.
Whoa there shill, remember all that money you’re being paid to spread the good word of Microsoft.
Btw, what is so special about the name “Nelson” at MS? I just don’t get it. “Major Nelson” is the main shill for the Xbox division and here we have a lowly professional bullshitter with the nick “Nelson” on the OSAlert forums.
How much work do you have to put in to get to the rank of “Major”.
Is that like a special name for the MS shill squad or what?
One day you’ll post something productive.
One day you’ll gain the “major” rank!
You want to show mobile users a (at least partially) different UI than desktop users. How do you detect “mobile”? Mobile in the sense I’m thinking of is a relatively small screen (tablets might have screens bigger (in virtual pixels) than old desktop PCs had!) and uses touch as primary input method. So you check for screen size? Well, there are netbooks etc. so no. You check for touch support? Well, desktop IE11 has touch support and with convertible handhelds you can expect that for more “desktop” browsers `”ontouchstart” in window` to yield true.
You don’t just want to present every browser with touch support the touch optimized UI, because people using IE11 might not even have a touchscreen! Maybe a combination of a relatively small screen and touch support would work, but I doubt it. And maybe tablets of the future will have really large screen sizes.
So what else is there? You can look at the user agent (you can even do that on the server). So looking for the string “Mobile” should work, right? Well, sadly not all mobile browsers put that string into their user agent. So I ended up writing a more sophisticated user agent detection. I look for “Mobile” or “Tablet” and for a number of known “tags” in mobile user agents.
Simplified excerpt (ua is the lowercase user agent string):
var browser = {
mobile: /(mobile|tablet|iphone|ipad|ipod|android|windows phone|blackberry|opera mini|webos)/.test(ua),
// also true on desktop IE11:
touch: “ontouchstart” in window || !!navigator.msMaxTouchPoints,
ios: /(iphone|ipad|ipod)/.test(ua),
…
};
I detect iOS so I can work around iOS specific bugs. I also have more complicated IE detection code (because IE11 does not contain “MSIE” in its user agent string). Yes, it’s horrible, but what else can I do? When possible I do feature detection, but things like “under iOS you cant use the lower 40px of your screen (no click/touch events)” is not detectable any other way.
Take a close look at the following link:
http://caniuse.com/#feat=css-gradients
Make sure you click the “show all” button. Now, move the cursor over the various versions and you’ll notice the following things:
1) Microsoft is always and consistently behind the other browsers in implementing CSS3 standardized things.
2) The features are prefixed for a long time with -webkit and -moz before standardized.
3) Microsoft virtually never implements a feature before it is standardized. And even then often a version or two later.
4) Microsoft releases browsers bound to operating systems and only once every 2 year or something. That means you sometimes have to wait years before the browser gets the feature.
5) There is no standard for detecting a mobile browser. They may mock some developer that tried to do detect it in some silly way, but fact is that there are only hacks for trying to detect the device. Also, notice how ‘mozilla’ is listed by ALL web browser vendors since 1865, so this problem has existed for a very very long time.
6) Websites are usually built, then only incrementally updated as needed. This means that if Internet Explorer doesn’t support a feature at the time the CSS file is being written, that is just too bad for IE. Likewise, if W3C hasn’t bothered to finish their standard in reasonable time, that too, is just too bad! The alternative is that none gets the enjoy gradients – is that what you’d prefer?
I think it is rather unfair to blame web developers that neither W3C or Microsoft are particular quick at adopting standards. Maybe if they did, web developers wouldn’t have to constantly prefix everything newer than CSS 2.1 (released ca. 1903)!
Uh, Microsoft supports more unprefixed CSS3 than any other vendor. Check it yourself.
I’m not talking about IE11 (they have been doing great progress lately), but the time period (last 5+ years of mobile) in which Thom blamed web developers.
Besides that, unprefixed CSS3 still means you are only tracking the things W3C *finally* standardized. That was half my complaint, that something has to be prefixed for ages and therefore increases the number of sites only using the prefix.
Historically, IE implemented features before they were standardized. This generated a lot of criticism and logistical problems when subsequent standards differed from existing implementations. One way of viewing this is the IE team listened to and accepted the criticism.
I’m not a web developer, but the ‘have you people learned nothing’ comment shows the lack of understanding that people have about the web. It’s a lovely idea to think that all web developers should write to open standards and it wouldn’t matter what web browser you used.
The reality is that the web is a platform, and a hugely complicated one. To make matters worse, it’s not no real compile time or runtime integrity, i.e. you don’t really know if something will fail until it fails. Then of course you’ve got lots of users using non-cutting edge browsers, a year or two old or older even.
That basically means that web sites can, and do fail *a lot*. Whether you write to open standards or not, your site is going to fail. So you can decide to hammer out the issues on all browsers, or just pick one.
Most people writing web sites aren’t doing it for the glory of HTML5 and CSS, they’re doing it because it’s their job, or otherwise to achieve a goal. That means working in the circumstances in which they’re in.
I’m not a web developer, but I’ve made websites in the past. In those days you wrote for IE compatibility, and while I rolled my eyes at that, I also was pleased that I only had to test on IE, not everything else.
The web is an incredibly complicated and fault-prone system, writing to one engine, WebKit is not good for the Internet, but it’s good for those making websites. By working to ensure full operation on all browsers, you’re making work for yourself and all you’ll be left with is a broken site on all browsers, rather than a chance at functionality on one of them.
It’s not about ‘learning nothing’, it’s about accepting the reality that the web is a total nightmare to work with, and this minimizes the nightmare.
Write to an older standard then, if you are writing to bleeding edge then you are destined to fail.
I’ve yet to see a site written to the older specs fail to render properly in anything but IE. Want to ensure it works there as well? Kick the Web2.0 stuff out and write even older, web classic if you will.
At least Webkit is free software. This means it is available on *every* single platform. With IE and its Trident engine, many people had zero way of accessing certain sites at all.
Jolla? FirefoxOS?
Jolla: http://linuxg.net/webcat-0-9-1-for-sailfish-os-has-been-released/
Edited 2014-08-03 02:57 UTC
Hm, one man show, linked news from January… (though I love the name
)
PS. And on the general note, Haiku also seems to have continuing problems with availability of current browser.
Depends what you mean by every single platform, if you mean ‘hardly any of them’, then yes, that’s correct. There are hundreds, if not thousands of computer Operating Systems and platforms. WebKit is available for a few of them. Open Source does not make something portable. You’ve still got have compatibility with the source. WebKit for example is mostly C++, if you don’t have a C++ compiler (or one which is new enough), then you don’t get a port. Same goes for the hundreds of dependencies WebKit has.
OpenOffice is Open Source, does not mean I can just build it on an Acorn RiscPC and it’ll work.
Sure, but then we’re talking about a baseline which is much lower already. It’s not just Webkit specific behavior which will prevent you from accessing a website. At least I’m not aware of any modern platforms that do have a modern web browser which attempts to support all recent standards, but do not have a C++ compiler.
Oh really ? If some system doesn’t have any decent/recent compiler available then it belongs in a garbage can and woudln’t be able to handle modern web anyway.
Edited 2014-08-03 13:39 UTC
I would argue that if a system has no modern compiler available, that would be a bigger issue than it not having a modern web browser.
Actually I was hoping for someone to come up with an example of a system which does have a relatively modern web browser, but no C++ compiler
Z_God,
A Java based OS would probably qualify, however to me this seems besides the point. In many cases it’s not the availability of webkit that’s the problem, it’s that the web shouldn’t be dependent on specific browsers in the first place. Once this happens, it may take a decade to recover from through natural obsolescence.
A solution would be for the browsers to go ahead and support the browser specific pages, but display a message to the user: “this page is rendered using proprietary standards.” This would attribute a bit of shame to the companies & web devs who didn’t use appropriate standards. But alas I can’t see apple doing this since non-standard pages designed for safari give them the upper hand.
Completely agree. I still believe that the current situation is less worse than what we had previously with IE, but yes it’s not good of course that people can’t manage to write complaint webpages.
It seems that a lot of people here do not understand how the modern web (HTML 5 + CSS 3) is actually being developed. There isn’t even an actual CSS 3 standard – just extensions standardized one at a time as an ongoing process.
It starts out with one of the browser vendors working on an idea for an improvement. Lets say webkit as an example, and they want to add gradients to backgrounds. Then, they add a -webkit extension that tests out this idea. At some point another vendor, lets say Mozilla, decides they like this initiative and they implement the same feature with a -moz prefix.
As more vendors now find interest in this new idea they create a w3c proposal describing this idea. The idea may turn out awful and get dropped, it might be redone several times, or it might finally (usually after years!) become a w3c recommendation (official standard). It is only then the very same feature now stops being a “proprietary” extension and gets a version without the prefix.
Web developers do not wait for the process to complete because some of the features are very much needed. Sometimes they are deal breakers (touch extensions on mobile devices, for example) for what they are trying to do. The process takes years, and in fact to some extent the web community uses the extensions as a feedback mechanism to test if a feature works as intended.
Showing some kind of alert because a feature is experimental would mean nobody could do touch specific web features for years after everyone bought a smart phone. The eventual standard would suffer if the feature isn’t tested in the wild before it is set in stone (standardized).
dpJudas,
Actually it’s not that we don’t know that, it’s just that when all is said and done it results in browser specific extensions on the web which don’t get fixed even when standards become finalized. Ideally extensions would be treated as experimental, and they would be fixed when the features gets standardized. Unfortunately many websites are developed once and then never get updated to comply with standards, they continue to exist for years. This is exactly why IE is choosing to render webkit’s extensions.
Even the W3C itself has considered adopting the webkit prefixes *as the standard* on the basis it’s the only way to make the standard interoperable with existing webkit deployments (browsers and websites).
http://www.sitepoint.com/w3c-css-webkit-prefix-crisis/
Imagine how much fun web development would be if that became standard practice! [/sarcasm]
The browser notification I mentioned earlier could help provide a motivation to become compliant.
Incompatibilities do exist beyond CSS as well, event driven ajax still has browser specific nuances (with state<4). In another instance I got called by a client reporting that a site navigation script didn’t work on the iphone, I had only tested it on android.
So I don’t know, maybe the desire to have one standard “HTML5” web which works everywhere without needing to test it everywhere is wishful thinking? Maybe what’s needed is a reference implementation – a browser who’s goal it is to work only with standards and zero non-standard features. All websites tested against this reference implementation would be known to work everywhere else.
I hate to point this out to you but OSAlert is using both moz and webkit extensions ( admittedly only a small number ).
Also, OSAlert has no mobile web experience – it doesn’t even scale properly – it has no ARIA support of any kind, and requires zooming in for navigation.
In contrast, Microsoft’s properties work great on iPhones, Androids, and Windows Phones.
Don’t you agree that a website that provides a good experience on all devices is the goal here, or are you after some kind of abstract puritanism where the web is mostly broken while we wait for a standards group to ratify every tag?
Perhaps before condemning all those developers who do, in fact, offer a rich accessible interface to mobile users, you should try to do that for OSAlert and see how you do without making use of extensions.
]{